This issue covers two weeks of healthcare 3D printing news, Nov 5 to 21. Jenny Chen asks who has the authority to say no in our industry, why hospital printing labs succeed or stall, and how an FDA rule on non-animal methods is turning human tissue models into paying contracts. Plus a patient-specific 3D-printed aneurysm clip, printed insoles for diabetic feet, and the week’s money and research.
The big thing
Who gets to say no?
In one week, the world’s most valuable chip company lost its founder, and the world’s most valuable AI lab went public. Even if you only care about the 3D universe, it is hard to be just a bystander.
I want to use that week to think out loud about three things I haven’t paid enough attention to: governance, the gap between a technology working and a technology being used (i.e., productivity), and how a regulator’s yes can create a market before the technology does.
1. Who gets to say no?
For thirty-three years, Jensen Huang was Nvidia. He co-founded it in 1993 and made it the supplier of nearly every chip that matters in AI. Success brings power, and power has a side effect: after enough years of being right, the people around you stop telling you when you might be wrong.
Nvidia’s board had no chair, and several “independent” directors had served since the 1990s. When reports surfaced of Huang’s relationship with a woman US officials suspected of ties to Chinese intelligence, the board first said it had confidence in him. Four days later he was gone. The board did say no, but only after the news was public.
I’ll admit I read more into the income statements than the proxy statements. What changed my mind was an episode of Prof G Markets, where Ed Elson sat down with NYU’s Aswath Damodaran, the Dean of Valuation, to go through Nvidia’s fall and Anthropic’s record IPO. His point: when so much of a company’s price rests on a few people, their judgment becomes part of the valuation. Anthropic’s seven co-founders hold 50.1% of the vote through one special share, and the stock is heading into the Nasdaq 100, which means your pension and my index fund. It’s worth the full listen.
Listen: Prof G Markets, “Aswath Damodaran On Nvidia’s Collapse & Anthropic’s $1.5T IPO” (Nov 20, 2026)
So what does this have to do with 3D printing?
Not much on the surface. Most of our OEMs are small cap businesses, commanding much less to no market or political attention. The large public players can be counted on two hands: Stratasys, 3D Systems, Materialise, and until recently Desktop Metal. Align Technology is a giant, but it makes an application, the aligner, not printers. In AI terms, Align is the app company, not the lab.
The OEMs are our industry’s labs. They build the general-purpose tool and bet that everyone else will find the use for it. That bet puts heavy weight on founders and boards: which applications to chase, when to raise, when to sell, and when to walk away from a deal. Our version is smaller and slower, but the structure is the same.
There’s a new trend worth watching. Tool makers increasingly are embedding themselves into the end application innovation cycle. For example, Adam Steege at Trio Labs invented a process that prints stainless-steel surgical instrument parts at five-micron resolution. Then he made an unusual choice: he doesn’t sell the printer. He sells the part. Device engineers want a part that meets spec, not a new machine to learn. Steege learned the cost of misalignment the hard way. His first medtech startup wound down, and it taught him that hospitals, insurers, investors and acquirers all have to agree before anything ships. (Disclosure: I’m an early investor in Trio Labs.) Listen: Episode #130
We also have our own governance history. Think of the poison pills and proxy fights between Stratasys and Nano Dimension, or Desktop Metal being absorbed by Nano Dimension and then filing for bankruptcy. Those weren’t technology failures. The printers worked. The decisions about ownership and control didn’t. Thirty years in, this industry has had plenty of brilliant founders and boardroom drama, but not the economic results its technology seemed to promise.
Hospitals face a smaller version of the same question. A point-of-care printing lab needs someone to decide who owns it (radiology, surgery or engineering), who signs off on a printed model, and who pays when reimbursement doesn’t. The programs that last answer those questions early. The ones that stall often have a great printer and no one with the authority to say yes.
I don’t think governance explains all of that, but I suspect it explains more than we admit. If you’ve been on a board, built a company or invested in this sector, hit reply. I’d love to hear what you think.
2. Technology doesn’t adopt itself
In 2016, one of the founders of deep learning said we should stop training radiologists because AI would be better within five years. Guess what, hundreds of companies were funded to replace radiologist, and we ended up with a graveyard of AI radiology startups, and ever more demand for radiologists now. This is not because the algorithms were bad, but because a hospital isn’t a benchmark dataset. It’s people, habits, liability, workflow and trust. The more successful AI companies now stopped trying to replace radiologists and started working with them.
3D printing is learning the same lesson. Last week I spoke with Dr. Mark Tan, a fellow neuroradiologist and Clinical Lead of the 3D Printing Centre at Singapore General Hospital. The program started with a surgeon’s printer at home and became a hospital department. It pays for itself: the center charges for the work and materials, because the hospital sees the value in more accurate surgery and shorter operating times. And it grew by word of mouth, surgeon to surgeon. Mark adapted what he learned from Stanford, Mayo and European centers to Singaporean medical system’s workflow, people and politics. The printer was never the hard part. Change management is. Listen: Episode #131
Two days earlier, at our virtual event on 3D-printed orthotics and prosthetics, five speakers from very different settings reached the same conclusion: the printer isn’t the bottleneck anymore. Expertise, workflow, clinical evidence and reimbursement are. One speaker estimated that only a quarter to a third of the O&P market has gone digital. Another is building an orthotics practice in Mongolia, which has no certified orthotists. The technology is ready. People and institutions are the slow part. Listen to the recording: Episode #132. Prefer to watch? Watch the highlights on YouTube.
3. The permission came before the market
For years, the pitch for human tissue models was simple: this will replace animal testing. The science was mostly there. The permission wasn’t. In September, the FDA published a rule that recognizes non-animal methods as a possible basis for safety decisions and removes wording that suggested animal studies were the only route. It also posted 25 real examples of how such methods have been used.
This week, buyers showed up. VivoSim Labs signed a seven-figure toxicology contract with a top-10 pharma company for its 3D human liver and intestine models. GATC Health and Hesperos will use linked organ-on-a-chip data to support an application to start human trials. Orakl Oncology raised $34 million to predict cancer drug response from patients’ own tumor organoids.
Notice the order. The technology didn’t create the market. A regulator saying yes did, and then the pharma toxicologists who carry the liability had to say yes too. Bioprinting companies selling into this space should take note. The customer isn’t buying a printer, or even a tissue. They are buying data a reviewer will accept.
So we are back to the first question: who gets to say no? Here, it’s the person who signs the filing.
From 3DHEALS
- Next-Gen 3D Printed Orthotics and Prosthetics: 2026 Event Recap: The takeaways from our Nov 12 event: the printer is no longer the bottleneck. Expertise, workflow, evidence and reimbursement are.
- 3D Printed Orthotics and Prosthetics: The 3DHEALS Guide: Our long-form guide to the technology, players, evidence and access gaps in printed O&P.
Regulatory watch
A quiet stretch: we found no new FDA 510(k) clearances for additively manufactured devices between Nov 5 and 21. The policy news, the FDA’s September rule on non-animal methods, is in The big thing above.
Clinical & research
- First functional patient-specific 3D printed aneurysm clip (World Neurosurgery, via PubMed): A Vienna team metal-printed custom clip blades from one patient’s imaging. In simulation, the custom design cut peak stress 14% vs a standard clip, and prototypes closed the aneurysm on a silicone model. In vitro, one case. DOI: 10.1016/j.wneu.2026.125528
- 3D printed insoles beat standard foam for diabetic feet (PLoS One, via PubMed): VA Puget Sound tested insoles with patient-specific shape and stiffness in 25 adults with diabetes. The printed designs lowered pressure on at-risk areas more than the standard-of-care insole. Lab walking trials, not ulcer outcomes. DOI: 10.1371/journal.pone.0363325
- Printed scaffold geometry cuts scarring around islet transplants (Acta Biomaterialia, via PubMed): University of Florida engineers found fibrosis rose with pore spacing. Their best design improved islet transplant outcomes in rats. A design lever for type 1 diabetes cell therapy. Preclinical. DOI: 10.1016/j.actbio.2026.11.019
Also this week
Companies and money
- Stratasys Q3 2026 results (Stratasys / SEC): Revenue $142.6M, up 4.1%. The CEO named dental among the businesses that grew; there is no separate medical figure.
- 3D Systems sets out its Formnext year (3D Systems): New MJP 300 printer (Q1 2027) and a stated focus on healthcare, dental and aerospace & defense.
- Precera Medical buys Additive Metal Services (Advanced Manufacturing): A medical device contract manufacturer adds metal injection molding and sinter-based AM for robotics, orthopedics and interventional parts.
- SINTX finishes its SiN/PEEK print facility (3DPrinting.com): Patient-specific implants from a silicon nitride/PEEK composite; product availability targeted for 2027, pending a 510(k).
- OSSTEC raises $8M for its 3D printed partial knee (VoxelMatters): The Imperial College spinout will fund the U.S. launch of its cementless Ascent knee.
- Lithoz upgrades its S65 Medical ceramic printers (VoxelMatters): A 4K projection pack doubles light intensity on the line used for bioceramic implants.
- Mitsui Chemicals invests in iORGANBIO (GlobeNewswire): AI and robotics to make more consistent human organoids for drug testing.
- bit.bio raises $25M (Tech Startups): More capacity for defined human cell types, the raw material of NAM and bioprinted models.
Research
- A $450 DIY bioprinter (University of Rochester / ACS Applied Polymer Materials): A converted plastic printer prints 100+ layers of living bacterial materials; build files are open.
- Bioprinted tumor microenvironment models (Seminars in Cancer Biology, via PubMed): A VoxCell-led review of bioprinted cancer models as alternatives to 2D culture and animals. DOI: 10.1016/j.semcancer.2026.102636
- A printed model of human lactation (Advanced Healthcare Materials, via PubMed): ETH Zurich prints breast-tissue scaffolds inside standard culture inserts in under 10 seconds each. DOI: 10.1002/adhm.71949
- Coaxial bioprinted hydrogel for cartilage repair (Advanced Materials, via PubMed): A Ningbo University sheath-core design shields stem cells from joint inflammation. DOI: 10.1002/adma.75685
- Recycled ABS passes cytotoxicity tests for printed guides (Bioengineer.org / Frontiers): Humanitas and Versalis printed anatomical models and surgical guides from chemically recycled plastic.
Know someone who should read this? Forward it, or subscribe to get The Lattice Brief every week.



Comments